Simulation and modeling for automation has advanced considerably. In one instance, manufacturers employ simulation for business purposes. While some have utilized simulation to close sales with suppliers, other manufacturers employ simulation for supply chain planning. For example, if it is known how many items are produced for a given line, then it can be determined where production needs to occur and what equipment needs to drive the production while yielding confidence in the final production outcome. Entities can also predict delivery schedules from simulations. Design engineers are using simulation to alter their designs to make products easier to manufacture, whereas many companies are now creating simulations of entire plants before a plant is built or refurbished.
One recent trend is the use of simulation to train plant personnel. There are two main areas where simulation has helped in training. In one, simulation allows less skilled workers to practice and gain experience “operating” plant equipment before taking the reins in the real world. In another, simulated operation offers an accelerated form of training. For instance, input/output (I/O) simulation software provides a shortcut to training on actual equipment that may not even be available at the present time, where training materials can be created from simulated manufacturing design. Training is often considered a secondary use of simulation, but the savings it produces can be considerable nonetheless. Another recent development in simulation mirrors progress in other areas of computer technology: standardization of data. One of the trends in simulation is the ability to share data. Thus, users share data in many directions, from product design and manufacturing to robot simulation and ergonomics, for example.
Three-dimensional modeling has gained ground in manufacturing simulation. Three-dimensional modeling first was applied in the aerospace and automotive sectors. Often, designers model robots in 3-D, then select the location for the respective operation such as “weld” and instruct the robot to perform along those lines. As for parameters such as pressure and the robot's maneuverability, such parameters can be built into the simulation and delivered by the robot manufacturer, thus preventing a simulation from inadvertently instructing the robot to perform an operation that is beyond its capabilities. Often times the robots are controlled from one or more programmable controllers that can also be simulated.
When a company has its manufacturing process fully simulated, it becomes easier to analyze a product design and observe how well it performs in a manufacturing setting. Since the design and manufacturing are not yet “live,” there is an opportunity to turn back to the design engineer and request changes before it is cost prohibitive to do so. Such changes at the simulation stage are generally much less costly to implement than at the actual manufacturing stage. Thus, early on in the life of the product, designers can analyze the simulated manufacturing process, and adjust a given product for desired manufacturability. The ability to alter a product design prior to manufacturing in order to cause the entire process to work more efficiently offers significant potential savings over the traditional design process. This process is often referred to as front-loading, where a designer can identify glitches in manufacturing through simulation and then facilitate planning on how to overcome such problems. With front-loading, products can be designed so it performs well in the manufacturing simulation which should mitigate problems in actual production thus mitigating overall system costs.
Simulation can also be implemented end-to-end, thus demonstrating how every process in a plant performs together over a designated period of time. For instance, simulation can occur from the IC (industrial controller) level up to warehouse management and other supervisory systems. One area where simulations of the entire plant are taking hold is with new plants or newly refitted plants. Before manufacturers determine what equipment they need and where it should go, they simulate the plant's entire operations. Dynamic simulation thus provides a model for a new plant to ensure the plant is designed properly.
It should be appreciated that such end-to-end simulations are complex. Several time scales and spatial scales associated with equipment, devices, and controllers are involved. The number of components that a simulation addresses grows as new elements of the industrial environment are incorporated; however, not only the size of the system increases but also its complexity through interactions of the added components. In addition, as industrial processes modernize and economical pressures drive development, simulations are sought as instruments for predicting performance, throughput, and cost of an industrial environment as a whole, from the bottom-up, starting at the device and controller level. In several instances, decision-making is based on simulation; therefore, predictive power and incorporation of the complexity of a simulated plant or factory.
With complex industrial control systems, there is a need to simulate the various components that makeup such systems. For example, this can include simulating multiple controllers over a network where each controller in turn controls subordinate I/O modules. Many times, multiple designers are involved where one designer will design one portion of the control system and other portions are implemented by other designers. In order to simulate an overall system model, there is a need to simulate and coordinate distributed components of the system.